Waiting to see someone we love can leave us restless and alert. So can waiting for an uncomfortable confrontation. The energy may feel similar, yet the experiences differ in ways that matter. Saying that we are worked up tells only part of the story.

Harold Schlosberg approached this problem by considering both the quality of emotional responses and their level of activation. His 1954 paper, Three Dimensions of Emotion, brought physiological research into conversation with studies of facial expressions. It offers an early way of thinking about feelings as variations along dimensions, while also showing why activation alone cannot tell us which emotion is present (Schlosberg, 1954).

What Is the Activation Theory of Emotion

A dimension is a continuum along which something can vary. Temperature, for example, changes by degrees rather than dividing neatly into separate kinds. A dimensional approach to emotion asks whether emotional responses can likewise be described through a small number of underlying variations.

Schlosberg considered activation a particularly useful way to understand the strength, or what he called the intensitive dimension, of emotional responding. He described a continuum extending from sleep through alert attention to highly activated emotional behavior. This placed strong emotion within a broader range of responsiveness rather than treating it as an entirely separate state (Schlosberg, 1954).

Schlosberg built on earlier work, including Elizabeth Duffy’s research on energy mobilization and Donald Lindsley’s activation theory. He brought this account of responsiveness together with his own research on the qualities conveyed by facial expressions. His three-dimensional proposal emerged from that integration (Schlosberg, 1954).

The distinction between activation and emotional quality gave each kind of evidence a purpose. Physiological measures could indicate changes in activation, while judgments of facial expressions offered a way to investigate differences among emotional responses.

The Three Dimensions of Emotion

Pleasantness and Emotional Valence

The first dimension concerns how pleasant or unpleasant a response appears. The related term valence describes this positive or negative quality in contemporary emotion research. The pleasure of receiving welcome news and the distress of receiving unwelcome news illustrate the difference.

Schlosberg’s immediate evidence came from observers rating photographs of posed facial expressions. Pleasantness–unpleasantness was one axis used to locate those expressions on a roughly circular surface. He cautiously suggested that dimensions useful for describing facial expressions might also help describe emotions more generally (Schlosberg, 1954).

Pleasantness describes how something feels, not whether the feeling is morally good or useful. An unpleasant emotion may still be understandable and worth attending to.

Attention and Rejection

Attention–rejection concerned the orientation suggested by a facial expression. Schlosberg described rejection as the active opposite of attention: lips, nostrils, and eyes compressed as though excluding an external object. He used these features to explain a rating dimension to observers (Schlosberg, 1954).

The dimension proved difficult to communicate through words alone. Ratings became more stable when observers were given example photographs showing the two extremes. These anchors gave judges a shared reference for interpreting the expressions.

Rejection has a particular meaning here. It refers to the exclusion suggested by an expression, rather than social rejection or a person’s decision to withdraw from a relationship. Attention–rejection also should not be treated as another name for the broader motivational distinction between approach and avoidance.

Activation from Sleep to Tension

The third proposed dimension concerns level of activation. Schlosberg used a sleep-to-tension rating scale when asking observers to judge the same facial photographs. Adding these judgments allowed him to sketch a three-dimensional arrangement of expressions (Schlosberg, 1954).

The additional ratings separated some expressions that occupied similar positions on the first two dimensions. Photographs representing grief, for example, received lower activation ratings than those representing pain and suffering. This result concerned the selected expressions; it does not establish how activated every grieving person will feel.

Schlosberg called the resulting figure a first approximation. The photographs emphasized emotional expressions and represented low activation states poorly. He wanted a broader sample, including sleep and low alertness, as well as skin-conductance readings from the people being photographed. Those measurements would help test the relationship between an expression’s rated activation and the person’s physiological state (Schlosberg, 1954).

Why Activation Alone Cannot Identify an Emotion

The reunion and confrontation in our opening example may both leave us keyed up. That shared energy does not tell us whether the anticipation is welcome or distressing. Understanding the difference requires information about the quality of the feeling and what it concerns.

Schlosberg recognized this limitation in activation theory. It addressed the strength of responding but did not adequately differentiate one emotion from another. Facial expressions offered an additional source of evidence about those qualitative differences (Schlosberg, 1954).

Schlosberg used skin conductance to track changes in activation. This measure records how easily a small electrical current passes through the skin, which changes with sweat-gland activity. A sudden sound or preparation for effort could change the reading, as could an emotional event. The measure could therefore indicate activation without identifying a particular emotion (Schlosberg, 1954).

Schlosberg considered skin conductance useful, while acknowledging that it was an imperfect indicator of a still-developing construct. He noted the importance of controlling conditions such as room temperature and electrode contact. His argument concerned measuring activation, not determining an emotion from a single bodily signal.

Schlosberg and the Circumplex Model of Affect

A circumplex arranges related qualities around a circle. Nearby positions represent greater similarity, while positions farther apart represent greater difference. In emotion research, this offers a way to show relationships among feelings that separate labels can obscure.

Schlosberg’s facial-expression surface used pleasantness–unpleasantness and attention–rejection. His 1954 proposal extended that surface by adding activation as a possible third dimension. James Russell’s 1980 circumplex instead organized affect in two dimensions: pleasure–displeasure and arousal–sleepiness. Russell’s arousal dimension therefore occupies a different place in his model from activation in Schlosberg’s three-dimensional proposal (Schlosberg, 1954; Russell, 1980).

Russell reviewed Schlosberg’s work alongside later studies in which observers did not clearly distinguish attention–rejection from sleep–tension. Those judgments suggested that the two scales could overlap. Russell’s account developed within this research history, using a different two-dimensional arrangement rather than keeping Schlosberg’s axes unchanged (Russell, 1980).

In Russell’s circumplex, excitement combines pleasant feeling with high arousal, while contentment occupies a pleasant, lower-arousal region. Distress lies on the unpleasant, higher-arousal side; a depressed feeling occupies an unpleasant, lower-arousal region. These are descriptions of affective qualities, not diagnostic categories (Russell, 1980).

Russell investigated the organization of affect through judgments about emotion words and self-reports of current feelings. This broadened the focus beyond Schlosberg’s posed-expression ratings. Although both models use dimensions and circular arrangements, their axes and the evidence used to develop them differ (Russell, 1980).

What Dimensional Models Help Us Notice

Dimensional descriptions can help us articulate differences that a broad emotion word leaves unstated. Pleasure may be lively and excited or quiet and settled. An unpleasant state may feel agitated or drained. These contrasts explain why asking only whether we feel good or bad can leave something important unspoken.

Russell later used the term core affect for the basic combination of feeling good or bad and feeling energized or tired. He proposed that we can experience such a feeling without knowing its cause, or connect it with something we believe caused it. This account belongs to Russell’s 2003 framework, not to Schlosberg’s original model (Russell, 2003).

For everyday reflection, interpreting emotion means considering more than whether we feel good or bad. We might notice how pleasant or unpleasant the moment feels, how energized or depleted we are, and what seems to be happening around us. These observations can make room for a more specific description.

Consider someone who calls a day bad. Feeling overstimulated after an argument differs from feeling flat after hours of monotonous work. Both experiences are unpleasant, but the difference in energy matters to what the person is trying to communicate. This example illustrates a use of dimensional language; the studies discussed here did not test it as a method of emotional regulation.

Limits of the Three Dimensional Model

Schlosberg’s evidence needs to be understood at the level at which it was gathered. His observers judged posed photographs. Their ratings described perceived expressions rather than establishing exactly what the photographed people felt. Schlosberg himself distinguished describing an expression from interpreting the circumstances that evoked it (Schlosberg, 1954).

The preliminary activation ratings also came from a restricted set of expressions, and the physiological measurements Schlosberg wanted had not yet been collected alongside those photographs. A promising arrangement of ratings was therefore an initial step toward testing a three-dimensional account.

Dimensional maps have a further limit: they cannot explain every difference among emotional episodes. Russell later acknowledged that pleasure and arousal alone did not adequately distinguish fear, jealousy, anger, and shame. His 2003 framework considered additional processes, including how a person interprets an event and attributes a feeling to its apparent cause (Russell, 2003).

A Few Words by Psychology Fanatic

We often reach for a single word when someone asks how we feel. Sometimes that word is enough. At other times, it hides the distinction between being peacefully content and happily stirred, or between feeling tense and feeling worn down. Taking a moment to describe those differences can make our experience easier to communicate.

When we are waiting for someone, a restless body may tell us that the moment matters. Understanding whether we feel glad, apprehensive, or conflicted takes us further into the situation. Schlosberg’s work makes room for that closer attention: we can notice the energy of an experience while still listening for what gives it meaning.

Associated Concepts

• Circumplex Model of Affect: Russell’s circular model organizes affect through pleasure and arousal, offering a comparison with Schlosberg’s proposed three dimensions (Russell, 1980).
• Emotional Valence: The pleasant or unpleasant quality of an experience is a dimension shared by these historical accounts (Schlosberg, 1954; Russell, 1980).
• Arousal Theory: These accounts examine differences in activation, including the sleep-to-alertness continuum discussed by Schlosberg (Schlosberg, 1954).
• Basic Emotions: Category-based accounts distinguish emotions such as fear and anger beyond their broad affective qualities (Russell, 2003).
• Cognitive Appraisal: Evaluating an event’s personal significance helps explain differences that pleasantness and activation leave unresolved (Russell, 2003).

Resources

Russell, James A. (1980). A Circumplex Model of Affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. DOI: 10.1037/h0077714.
(Return to Main Text)

Russell, James A. (2003). Core Affect and the Psychological Construction of Emotion. Psychological Review, 110(1), 145–172. DOI: 10.1037/0033-295X.110.1.145.
(Return to Main Text)

Schlosberg, Harold (1954). Three Dimensions of Emotion. Psychological Review, 61(2), 81–88. DOI: 10.1037/h0054570.
(Return to Main Text)